cs.LG（2025-05-14）

📊 共 16 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (9 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱八：物理动画 (Physics-based Animation) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#题目一句话要点标签🔗⭐
1 Emotion Knowledge Enhancement for Vision Large Language Models: A Self-Verification Approach for High-Quality Emotion Instruction Data Generation 提出SEKE框架，利用自验证方法增强视觉大语言模型的情感知识，并生成高质量情感指令数据。 large language model
2 Towards Fair In-Context Learning with Tabular Foundation Models 针对表格数据ICL，提出不确定性采样方法，提升TabPFNv2等模型的公平性。 foundation model ✅
3 Analog Foundation Models 提出一种通用方法以适应模拟内存计算的LLM foundation model ✅
4 LAS: Loss-less ANN-SNN Conversion for Fully Spike-Driven Large Language Models LAS：用于全脉冲驱动大语言模型的无损ANN-SNN转换 large language model ✅
5 Lossless Compression for LLM Tensor Incremental Snapshots 提出LMC：一种基于字节分组和Huffman编码的LLM张量增量快照无损压缩方法 large language model
6 Adversarial Suffix Filtering: a Defense Pipeline for LLMs 提出对抗后缀过滤（ASF）防御管线，有效抵御大语言模型的对抗攻击。 large language model
7 Depth-Based Local Center Clustering: A Framework for Handling Different Clustering Scenarios 提出基于深度局部中心聚类(DLCC)框架，以应对不同聚类场景的挑战。 multimodal
8 Layered Unlearning for Adversarial Relearning 提出分层卸载（LU）算法，提升语言模型对抗性重学习的鲁棒性 large language model
9 CXMArena: Unified Dataset to benchmark performance in realistic CXM Scenarios CXMArena：统一数据集，用于评估LLM在真实客户体验管理场景中的性能 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
10 A Multi-Task Foundation Model for Wireless Channel Representation Using Contrastive and Masked Autoencoder Learning 提出ContraWiMAE，一种用于无线信道表征的对比掩码自编码器基础模型 representation learning masked autoencoder contrastive learning
11 Adversarial Attack on Large Language Models using Exponentiated Gradient Descent 提出基于指数梯度下降的对抗攻击方法，有效破解大型语言模型。 reinforcement learning RLHF large language model ✅
12 Reinforcement Learning for Individual Optimal Policy from Heterogeneous Data 提出P4L算法，解决异构数据下的个体最优离线强化学习问题 reinforcement learning policy learning offline RL
13 Community-based Multi-Agent Reinforcement Learning with Transfer and Active Exploration 提出基于社群的多智能体强化学习框架，实现知识迁移和主动探索 reinforcement learning
14 Neural Multivariate Regression: Qualitative Insights from the Unconstrained Feature Model 利用无约束特征模型分析神经网络多元回归，为模仿学习等任务提供设计指导。 reinforcement learning imitation learning

🔬 支柱八：物理动画 (Physics-based Animation) (2 篇)

#题目一句话要点标签🔗⭐
15 Neural models for prediction of spatially patterned phase transitions: methods and challenges 利用神经网络预测空间模式相变，揭示早期预警信号的局限与泛化能力。 spatiotemporal
16 Generating time-consistent dynamics with discriminator-guided image diffusion models 提出时间一致性判别器，引导预训练图像扩散模型生成时序动态 spatiotemporal

⬅️ 返回 cs.LG 首页 · 🏠 返回主页